Osteoarthritis (OA) is a prevalent and costly disease, the early stages of which are extremely difficult to study in human beings. OA, closely resembling the human disease, occurs naturally in the knee joints of cynomolgus monkeys (Macaca fascicularis) and, similar to the human disease, increases in severity with age. The broad, long-term objective of these studies is to validate methods that may be used in vivo to diagnose accurately the early articular cartilage and bone changes in OA of the knee joint in this unique animal model. This will allow us to do future in vivo longitudinal studies to assess the effects of diet and hormonal and pharmacological therapies on the progression of this disease. The specific aims are: 1) to determine whether synovial fluid levels of proteoglycan epitopes that are anabolic (recognized by monoclonal antibodies 3B3 and 7D4) or catabolic (recognized by 5D4, BC3, and BC4) markers of proteoglycan metabolism are representative of OA markers of severity; 2) to determine whether synovial fluid levels of intact and degraded cartilage oligomeric matrix protein (COMP) are representative of OA severity; and 3) to determine whether micromagnetic resonance imaging (microMRI) is an effective method for determining the severity of bone and articular cartilage changes in this model. These aims will be achieved by comparing radiographic (conventional as well as serial section microradiography), histological, and immunohistochemical findings in knee joints from cynomolgus monkeys with a wide range of OA severities with levels of markers in synovial fluid and microMRI images from these same joints. Knee joints from 300 monkeys from the cynomolgus monkey colony or that were subjects in long-term clinical trials will be examined and graded morphologically to characterize the lesions of OA, and the data will be summarized using factor analysis. Synovial fluid assays will be done using established techniques (ELISA, radioimmunoassay, or Western blotting) and techniques for microMRI will be modifications of methods used in human beings. The derived factor scores (from the morphological data) and the synovial fluid data will be used in correlation analyses to determine whether there is a relationship between synovial fluid levels of marker proteins and OA severity. Quantitative and qualitative evaluations of microMRI scans will similarly be compared with OA severity data to determine if this technique can be used to diagnose accurately early lesions of OA.